Pesticides are chemicals designed to cause biologic harm to plants, insects, and other pests. Widespread use in agriculture and in urban areas creates an opportunity for exposure to a large segment of the population. Epidemiologic investigations are undertaken to identify and clarify cancer risks from pesticide exposure. The Agricultural Health Study is a major effort which is designed to evaluate the relationship between pesticides and other agricultural factors and the risk of cancer and other diseases. Enrollment in this study includes about 90,000 men and women from Iowa and North Carolina. From 1993-1997 detailed information was obtained by self-administered questionnaire and in the next three years this information is being updated and expanded by telephone interview. Telephone interviews include questions on pesticides used and other agricultural exposures, lifestyle factors, medical and family history of disease, and diet. We are also collecting buccal cell samples as a source of genomic DNA on all members of the cohort. Analyses on cancer incidence and other chronic diseases is just getting underway with a focus on cancer incidence overall, prostate cancer and breast cancer during the next year. Preliminary findings from the prostate cancer anlysis suggests a number of pesticides are associated with a prostate cancer risk. The risk of prostate cancer is particularly strong among those with a family history of prostate cancer. In a case-control study nested with the Agricultural Health Study, retinal degeneration was significantly associated with exposure to fungicides and selected insecticides, further analyses is now being conducted on this finding. The risk of a high pesticide exposure event resulting in symptoms is more pronouned among farmers whose farms were in poor financial condition compared to those whose farms were not and among those with a higher risk acceptance than those with a lower risk acceptance. An analysis of mechanical injury associated operating a farm has just been completed and should be submitted for publication during 2001. A study of Parkinson's disease nested within the Agricultural Health Study is now being pilot tested. Several other projects to evaluate the cancer risk from pesticide exposure are underway. During the past year, analysis of data from case-control studies of non-Hodgkin's lymphoma noted an association with agricultural use of lindane, but not with DDT. However another case-control study indicated that chemotherapy for non-Hodgkins Lymphoma affects blood levels of organochlorines, which could result in misclassification of exposure and distort findings. Farmers were found to be at elevated risk for prostate cancer in a study of mortality in 24 states. Contrary to earlier investigations, however, the risk of breast cancer was not associated with serum levels of DDT among women in Mexico City. A methodologic study to compare pesticide levels in dust samples from vacuum cleaner bags with levels from a high-volume surface sampler (HVSS) found a high degree of correlation. This suggests that collection of vacuum cleaner bags, which is a simpler and less expense approach than the HVSS, provides a reliable indication of pesticide levels on dust in homes. Several feasibility studies have been conducted among migrant and seasonal farm workers to determine appropriate methods for epidemiologic investigations in this understudied group. Results from this effort appear in a special volume of the American Journal of Industrial Medicine in the near future. A review of the literature was conducted which suggested that pesticides may play a role in the development of childhood and adult cancers. Recently completed case-control studies of cancers of the brain and stomach in Nebraska will provide the opportunity to assess the role of agricultural, general environmental, and lifestyle factors in the development of these tumors. Blood and urine have been collected from 30 applicators applying the herbicide 2,4-D to relate dose to possible biologic effects, and dose-response gradient was found for 2,4D and mitotic index. In a separate study, preliminary results suggest that farm exposures are more strongly associated with NHL with a t(14,18) translocation than other types of NHL. In yet another study, preliminary results suggest that maternal occupational exposure to agricultural chemicals may increase the risk of giving birth to a child with limb defects.